Asynchronous Motor Fault Detection And Diagnosis Based On Matrix Pencil Arithmetic

With the continuous development of science and technology and the on-going progress of scientific level,the application of asynchronous motor becomes more and more extensive.Asynchronous motors have become an irreplaceable driving device in various industries and have a large variety and large numbers,meanwhile,the types of machine failures of asynchronous motors also increase.The asynchronous motors can diagnose the failures of motors timely,accurately and effectively to avoid serious accidents,which has great significance for industrial production.This paper takes different types of failures of asynchronous motor as the research project,based on which Matrix Pencil algorithm is proposed.The Matrix Pencil algorithm is the reconstruction of the asynchronous motor and we use it as Hankel matrix,extract the singular value of Hankel matrix,arrange it indepreciating and descending order,partially zero processing,determine the order,eliminate the signal noise and extract the signal characteristic parameters.In this paper,we simulate the stator single-phase windings short circuit and Rotor broken bar with different types and degrees of failures respectively according to the Matlab simulation platform.By applicating the comparative analysis of the spectrum simulation diagrams before and after,we can prove that the Matrix Pencil algorithm can effectively suppress noise,extract the fault feature component accurately and provide effective methods for detection and fault diagnosis of asynchronous motor and set up the FPGA hardware circuit.According to the types of asynchronous motor failures,we can confirm the asynchronous motor fault diagnosis system.In order to process the signal data efficiently,quickly and accurately,this paper selects FPGA as the controller of failure diagnosis based on Matlab simulation and verification algorithm.The motor current signal conversion is completed by conditioning circuit and high-speed ADC module and transmitted to the FPGA control module for Matrix Pencil algorithm program processing.SDRAM storage module is used to store data information processed by Matrix Pencil algorithm,and Ethernet communication module is used to transmit data to the upper computer display to have a real-time detection and diagnosis of asynchronous motor failure condition.Based on the hardware system of fault diagnosis,the application of Matrix Pencil algorithm on stator single-phase winding short circuit and rotor broken bar failure diagnosis in asynchronous motor is implemented.We use testbench test file as an excitation under the Modelsim software platform to complete the output current spectrum and the output current and power spectrum analysis of broken bar fault of different degrees of short circuit fault between stator single-phase failure..It can be concluded that Matrix Pencil algorithm can effectively eliminate noise interference,extract fault feature information in asynchronous motor fault diagnosis and achieve the realization of asynchronous motor fault detection and diagnosis based on Matrix Pencil.